The Initial Impact of COVID-19 on Labor Market Outcomes Across Groups and the Potential for Permanent Scarring

ESSAY 2020-16 | JULY 2020

Betsey Stevenson

i The Hamilton Project • Brookings MISSION STATEMENT

The Hamilton Project seeks to advance America’s promise of opportunity, prosperity, and growth. The Project’s economic strategy reflects a judgment that long-term prosperity is best achieved by fostering economic growth and broad participation in that growth, by enhancing individual economic security, and by embracing a role for effective government in making needed public investments. We believe that today’s increasingly competitive global economy requires public policy ideas commensurate with the challenges of the 21st century. Our strategy calls for combining increased public investments in key growth-enhancing areas, a secure social safety net, and fiscal discipline. In that framework, the Project puts forward innovative proposals from leading economic thinkers — based on credible evidence and experience, not ideology or doctrine — to introduce new and effective policy options into the national debate.

The Project is named after Alexander Hamilton, the nation’s first treasury secretary, who laid the foundation for the modern American economy. Consistent with the guiding principles of the Project, Hamilton stood for sound fiscal policy, believed that broad-based opportunity for advancement would drive American economic growth, and recognized that “prudent aids and encouragements on the part of government” are necessary to enhance and guide market forces.

ii The Hamilton Project • Brookings The Initial Impact of COVID-19 on Labor Market Outcomes Across Groups and the Potential for Permanent Scarring

Betsey Stevenson University of Michigan

JULY 2020

This policy essay is an essay from the author(s). As emphasized in The Hamilton Project’s original strategy paper, the Project was designed in part to provide a forum for leading thinkers across the nation to put forward innovative and potentially important economic policy ideas that share the Project’s broad goals of promoting economic growth, broad-based participation in growth, and economic security. The author(s) are invited to express their own ideas in policy papers, whether or not the Project’s staff or advisory council agrees with the specific proposals. This policy paper is offered in that spirit. The author(s) did not receive financial support from any firm or person with a financial or political interest in this article. They are currently not an officer, director, or board member of any organization with an interest in this article. at this realization do they shed workers. Across the economy Introduction the process of businesses reducing hiring or closing up shop altogether, as well as fewer businesses being created, can The arrival of the novel coronavirus in the United States often last for years, with job losses accumulating over time. brought with it a public health crisis that meant that The labor market becomes like a game of musical chairs in previously advantageous ways of organizing work and home which more chairs are removed each month and those sitting life carried new costs. As COVID-19 (the disease caused by in chairs increasingly refuse to stand up and potentially free a the novel coronavirus; hereafter COVID) began to spread chair for someone else. As a result, long-term unemployment in early March, schools closed and businesses sent workers grows over time, and labor force participation falls slowly as home as all but essential services temporarily shuttered. More unemployed workers give up and leave the labor force and than 31 million applications for unemployment insurance those considering whether to enter or reenter the labor market were filed during March and April, a period during which become discouraged about the prospect of finding work. the economy was essentially put on pause and many jobs were temporarily suspended.1 The official unemployment During the last recession, job growth slowed in 2007, before rate peaked at 14.7 percent, but the logistical challenges with consistent job loss began in February 2008, a month when measuring a surge in people who were suddenly laid off means the economy lost 79,000 jobs. Monthly job loss accelerated that, more realistically, the unemployment rate might have over the next year, hitting a peak monthly loss of 800,000 been over 20 percent at its peak.2 The need to stay home in jobs in March 2009. Job losses continued through mid-2010, order to stay safe caused many people who were not currently by which point 8 million jobs had been lost. Even after job in the labor force, but who would have started looking for a growth slowly resumed, long-term unemployment and job in March and April, to put job-finding plans on pause. declines in labor force participation continued for years. The data show that new entrants and reentrants to the labor The prime-age labor force participation rate only began to force plummeted both in absolute numbers and as a share of consistently improve at the end of 2015 (Breitwieser, Nunn, the unemployed. Labor force participation fell to 60.2 percent and Shambaugh 2018). in April 2020, a low last seen in the early 1970s. In contrast, over the two months of March and April 2020, The sharp declines in spending and work were guided by both the overall labor force participation rate and the state stay-at-home orders, although research has shown that prime-age participation rate fell far below the low rate that the driving force was voluntary as many people stayed away the previous recession took 5 years to hit. And yet, nearly from crowds and businesses began to implement work-from- half of the decline had been reversed by June as labor force home policies prior to state policies being enacted.3 Those participation rose in both May and June. Similarly, the who could do so worked from the safety of their homes, while unemployment rate hit a high not seen since the Great others were temporarily laid off waiting to find out when Depression just two months into the recession, and has since and if they would be called back to work. Many of the self- reversed about 40 percent of its climb. employed found that they were no longer able to provide their services and were left, along with other business owners, to However, the traditional benchmarks measured in April reexamine their business models to assess how likely it is that had captured neither the damage to the labor market nor they will be able to resume pre-pandemic operations in the the permanent changes in workers’ attachment to the labor medium term. force. Both unemployment and the decline in labor force participation reflected many truly temporary layoffs since Parents faced an added challenge as those who had previously some workers who were sent home would be needed back relied on schools and child-care centers were forced to in their jobs as soon as the economy could reopen. But take on new roles as educators and round-the-clock child- even as millions have returned to work, both the employers care providers as schools and child-care centers closed. who did and those who did not originally do temporary The pandemic also interrupted alternative forms of child layoffs continue to grapple with how to adjust to a changing care, including relying on older family members such as economic and public health situation. Permanent job loss and grandparents. Many working parents were left scrambling to worker detachment from the labor force is occurring slowly, find a way to simultaneously do their job and care for their with each day bringing new layoffs that no longer represent a children. business pausing, but rather are increasingly likely to reflect a business reorganizing or closing. The job loss and unemployment witnessed early on was unlike a normal recession. In a normal recession, it takes time Demand in the economy, and therefore labor demand, was for employers to realize that demand for their product has lowered both directly and indirectly from the pandemic. The declined or that their business model is not robust enough to direct effect is a suppression—people are staying home or be sustainable in a weak economy. Only once employers arrive avoiding certain types of spending to avoid being infected

2 The Hamilton Project • Brookings by the virus (or because state stay-at-home policies and other jobs have accounted for more than half of the growth in restrictions have constrained them). But labor demand is also nonfarm payroll jobs held by women in the 21st century. being affected by overall concerns about current and future Jobs in education and health services have grown nearly income and the economy. Only once the labor market is no continuously over time including through past recessions. longer being suppressed by both stay-at-home measures and Job loss began in March, but primarily occurred in April, at people’s actions to directly avoid viral infection will we learn which point employment had declined by 11 percent. Women the unemployment and labor force participation exit rates made up 82 percent of the lost jobs in that industry. from which we must slowly recover. While these two sectors help explain why women got hit A closer look at how job loss unfolded and recovered across hardest as the pandemic began, they also explain why the different groups provides some insight into what the future gender gap in unemployment and labor force participation may hold for the labor market. Equally, it is important to rates narrowed substantially in June as many of these jobs realize that no one is able to assess the long-term scarring recovered. By June, employment in leisure and hospitality had that will occur in the labor market until the pandemic is partially rebounded, closing one third of the gap. Similarly, closer to being resolved. Major sectoral shifts in the workforce roughly one quarter of the lost jobs in education and health will likely be necessary, and while many of those who were services recovered as medical and dental practices began to hurt initially during the shutdown will bear the brunt of resume seeing patients for more than just urgent care. this reallocation, its permanent impact will likely be more narrowly targeted.4 While the initial rebound has helped narrow the gender gap in unemployment and labor force participation, women’s dominance as the majority holder of nonfarm payroll jobs is A Gendered Shut Down unlikely to recover any time soon. The pandemic has also hit women harder than men by the In December 2019, women achieved a milestone in the labor increased burden of care since children’s schools, daycare market: they held more nonfarm payroll jobs than men. providers, and camps have closed, and many remain closed. While it was not the first time that women had outnumbered Additionally, many families have had to consider how to men in the labor market, it was the first time that women had best provide elder care and how to ensure the safety of those overtaken men during a period of job growth.5 January and more vulnerable to the worst effects of COVID. Women’s February 2020 continued this trend as women’s slight edge traditional caregiving role and the crisis of care that many in terms of nonfarm payrolls continued, and at the start of families are facing in the United States could have long-term the year it seemed likely that women would continue in this repercussions for women’s labor force attachment and success, dominant position far into the future. But women’s advantage although we have yet to see this impact in the data. Women was rapidly undone by pandemic job loss, and by May 2020 with children under age 18 at home were no more likely than women held only 49.2 percent of nonfarm payroll jobs. other women to leave employment this spring. However, with so few people having already returned to work, it is unclear Men typically lose work early in a recession because they tend whether child-care issues will prevent women from returning 6 to be employed in industries that are more cyclical. However, to employment as the economy recovers. the early stage of the pandemic saw rapid declines in female- dominated industries. For example, the most rapid declines More generally, the way in which increased caregiving in March were in employment in leisure and hospitality, typically affects women’s labor market outcomes occurs an industry in which 53 percent of workers were female in slowly over time. Caregiving needs lead women to choose February. Cuts in this sector made up more than half of the jobs with more flexibility and shorter commutes.7 They may decline in March 2020 nonfarm payrolls, and women held switch to part-time work or step out of the labor market for 57 percent of the jobs cut. Job loss in leisure and hospitality a period. The repercussions of these choices are that women’s was 10 times larger in April, and once again women made up wages tend to grow more slowly because women take and are a disproportionate share of the jobs cut. By April, nearly half given fewer opportunities for promotions and higher wages. the jobs in this sector had disappeared from payrolls. The impact of the difficult choices that woman have made and will continue to make because of the lack of child care due A similar pattern played out to a lesser extreme in education to the pandemic will likely affect these women’s labor market and health services, although this is an industry in which outcomes for decades. 78 percent of the workers in February were women. These

3 The Hamilton Project • Brookings as seen in figure 1. These declines come after a period of New and Reinforced Labor Market rising labor force participation among Hispanic and Black Inequalities workers. The historic gap in which White workers tended to have higher labor force participation than Black workers has Low-wage workers have borne the brunt of the economic reemerged after having been nearly eliminated at the start of pain of the pandemic. Because low-wage work is more likely the pandemic. to involve in-person tasks, those workers faced a double hit from the pandemic. Many low-wage workers were laid off as in-person work slowed and then stopped for nearly all but A Snapshot of Those Employed essential employees. Those who were kept on the job—often as essential employees—faced new health risks for which Prior to the Pandemic there was little additional compensation. A more complete picture of what has happened to workers Similar to the patterns for women, low-wage workers are who were employed prior to the pandemic and of the concentrated in leisure and hospitality, which is the sector disparities in the labor market emerges by following the labor with the lowest hourly earnings. Low-wage workers are also market path of those who were employed in February.8 As concentrated in retail trade, which contracted by 15 percent people left jobs they held in February, some transitioned to overall. As a result, unemployment among workers with unemployment, others exited the labor force, and still others less education and among Black and Hispanic workers were employed but absent from work. The many different skyrocketed. The unemployment rate of those with only a paths out of employment that workers took in March, April, high school diploma rose from 3.6 percent in February to and May highlights the incomplete nature of measured 17.3 percent in April, while the rate for those with at least a unemployment. four-year college degree went from 1.9 percent to 8.4 percent. The unemployment rate among Hispanic workers rose to Overall, more than one in four of those employed in February 18.9 percent, and for the first time exceeded that of Black had a spell of nonemployment by May, meaning that they had workers, whose unemployment rate hit 16.7 percent. either been officially counted as unemployed, not in the labor force, or employed but absent from work for other reasons Similar trends were seen in labor force participation, which (Bureau of Labor Statistics [BLS] 2020b; author’s calculations). fell more for those with less education. Looking along racial It is not clear how we should code people who were employed lines, the labor force participation rate among Hispanic but absent from work. Some of these workers were on vacation workers fell the most, followed by that of Black workers. In or family-related leave that was not influenced by COVID. But each case, labor force participation fell by substantially more the large spike in people who told interviewers that they still than it fell during the Great Recession, and hit modern lows,

FIGURE 1. Labor Force Participation Rates by Race and Ethnicity, 2000–20

75 ) t e n

c 70 e r p ( Hispanic e c r o f

r 65 o

b White, non- Hispanic a l

f o e r

a 60 h

S Black, non- Hispanic

55 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020

Source: Bureau of Labor Statistics (BLS) 2000–20. Note: Shaded bars refer to recessions.

4 The Hamilton Project • Brookings had a job but could not go to work has suggested that many of bachelor’s degree experienced a spell of nonemployment in these people are perhaps more accurately described as “laid March, April, or May, compared to 18 percent of those with at off.” More generally, it is clear that COVID-related causes led least a college degree. The biggest disparity occurred in April to an increase in workers being absent from work.9 Similarly, when nearly 3 in 10 workers without a bachelor’s degree who some workers who were counted as not in the labor force, had been employed in February were not employed. because they were not available to take a job or because they had not looked for work, would have been either employed Hispanic and Black workers were more likely to experience or officially unemployed had the pandemic not hit. As such, a spell of nonemployment, with about one in three who a more complete picture of how the pandemic affected were employed in February experiencing a month or more the labor market should include at least some of those who of nonemployment. Nonemployment among those who had have a spell out of the labor force. Taking all these forms of been employed in February peaked in April, when 29 percent, nonemployment together, roughly twice as many people who 27 percent, and 20 percent of Hispanic, Black, and White were employed just before the pandemic hit experienced a workers, respectively, were not employed. spell of nonemployment (30 percent) compared to measuring 10 Looking across all groups of workers, the statistics in table 1 only spells of unemployment (16 percent). show that the nonemployment rate among those employed Workers without an undergraduate degree experienced in February remained near April rates in May. Table 1 some of the highest rates of nonemployment, just as they shows that the nonemployment rate among Black workers experienced a greater increase in unemployment and a steeper who had been employed in February remained just below fall in labor force participation; they were also more likely to 27 percent—the April high point—in May (26 percent), while be coded as employed but absent from work for other reasons. the unemployment rate among White and Hispanic workers Table 1 shows that 35 percent of workers without at least a who were employed in February declined slightly. Turning

TABLE 1. Labor Market Outcomes for Those Employed in February by Selected Demographic Characteristics At least Continously At least one month Nonemployment employed At least one one month employed but February month of not in the absent from through May unemployment labor force work March April May Percent Gender Men 75.2 14.5 9.3 6.4 5.3 20.1 17.0 Women 68.8 17.8 12.4 8.1 7.2 25.5 22.2 Race and ethnicity White, non-Hispanic 75.7 13.6 9.3 6.4 5.2 19.8 16.6 Black, non-Hispanic 65.9 19.8 14.1 8.6 7.3 26.8 25.7 Hispanic 64.5 22.3 13.3 8.3 8.5 29.2 24.4 Educational attainment Less than a bachelor’s degree 65.5 20.6 13.6 8.4 7.7 28.8 24.0 Bachelor’s degree or more 81.6 9.7 6.8 5.5 4.2 14.0 13.1 Presence of a child Women without a child 67.4 18.7 13.6 7.7 7.9 27.7 23.5 Women with a child 70.4 16.8 10.9 8.6 6.4 22.9 20.7

5 The Hamilton Project • Brookings FIGURE 2. Labor Market Transitions for Those Employed in February, Selected Demographic Characteristics Nonemployed in Nonemployed Nonemployed in Nonemployed in Employed in May, May, employed in for all three May and April, May, employed in nonemployed in April, nonemployed months employed in March April and March March or April in March

Women with a child Women without child Presence of a child Bachelor's degree or more Less than a bachelor's degree Education Hispanic e c

a Black, non-Hispanic R White, non-Hispanic r

e Women d n e

G Men

0 10 20 30 40 Share of group that was employed in February (percent)

Source: Bureau of Labor Statistics (BLS) 2020b; author’s calculations. Note: Data are for 2020. The author analyzes the one-quarter of observations that can be successfully linked through each of the 4 months (February, March, April, and May) in the Current Population Survey. “Not employed” refers to those who are employed but absent from work for other reasons, unemployed, and not in the labor force. back to differences by educational attainment, there was a nonemployment rate among those with a bachelor’s degree Conclusion in May that was similar to the rate in April, whereas the nonemployment rate had improved slightly among those with The job loss that we have witnessed so far has largely been less education. temporary, but permanent job loss is rising. In June, although net job growth reflected nearly 5 million people returning to Figure 2 shows that this reflects a large share of workers who jobs, permanent job losers rose to 2.9 million. While it will had not been brought back to work by May: the light orange likely be years before we can excavate from the data we are bars represent the share that were reemployed in May. The now observing the permanent damage the pandemic has gray bars show those whose first spell of unemployment was done to the economy, the acceleration in permanent job loss is in May, partially offsetting those who had been brought back an indicator that the damage to workers’ connections to their to work. jobs and the labor market may continue to occur for many more months. These data highlight that the short-term challenges that workers have faced were greatest among the most vulnerable Even those who continued working have found that their workers. Few of those who lost work have returned to jobs have changed.11 New technology was thrust upon employment, and new workers continue to experience job workers who had to learn how to do virtually what they loss. The sheer magnitude of the gaps—nonemployment once did in person. This unprecedented surge in the use rates that remain nearly 10 percentage points higher both of technology to accomplish tasks that used to be done in for minorities than for White, non-Hispanic workers and for person has permanently changed people’s knowledge of, and those workers with less education compared to those with a comfort with, these substitutes. As a result, it will likely cause bachelor’s degree or more—suggest that it is minorities and permanent changes in how people work and the technology workers with less education who are most at risk of being left that they use. These changes have yet to fully percolate behind as the economy recovers. through the labor market to affect staffing decisions. Will medical practices continue with virtual appointments long after COVID has passed? Will business meetings that once required travel become Zoom meetings? These changes will ultimately lead to a group of workers who will need assistance

6 The Hamilton Project • Brookings in transitioning to new jobs, such as through expanded job the pandemic. In the long run, the American economy must training and job search assistance programs. build better on-ramps to employment for those who take a period out of the labor market if the economy is to reach its While Congress has scrambled to save airlines on the belief full potential. that air travel is essential for a well-functioning modern economy, they have overlooked what is perhaps the most Finally, as we approach the middle of the summer with important industry in a modern economy: our child-care each day bringing a record high of new COVID cases, it is providers and schools. Parents will continue to struggle with clear that millions of workers will need to continue to rely child-care issues, particularly with the potential of children on income support. The CARES Act expanded both the out of school and without child care this coming fall and number of people covered by unemployment insurance and the risk to grandparents of relying on them for child care. the amount of income they receive. Yet, at the end of July, The pandemic has highlighted the fact that child care is not unemployment insurance payment amounts will fall by a women’s issue, it is not a personal issue, it is an economic more than half as they return to state-determined payment issue; parents cannot fully return to work until they are able amounts. Expanded benefits provided so far have clearly to ensure that their children can safely return to child-care helped bolster the economy: household incomes rose in April and educational arrangements. The child-care crisis spurred and, in response, retail sales rose strongly in May. However, by the pandemic could force families to make difficult consumer spending has yet to return to pre-pandemic levels, decisions that will lead to lower labor force participation and and many fear a wave of evictions as landlords expect out- lower earnings for decades to come. of-work individuals to resume paying rent. In order to foster a return to work, we need to support a continued return to The solution to preventing large-scale permanent scarring, spending. particularly among women, is to prioritize safely opening schools, to ensure that child-care centers do not go bankrupt The lifting of state shut-down orders ended the initial and that the centers have the resources to adapt their widespread suppression of the labor market. The question buildings and practices to new protocols like improved air now is how the labor market will be rebuilt. In its wake, there flow and increased surface disinfecting, and to encourage is likely to be a massive reallocation in who works, how they workplace flexibility. In addition, job-protected paid sick work, and what kinds of jobs they do. Since April, we have leave as well as medical and family leave are more important seen the official unemployment rate fall to around 11 percent, than ever. People will continue to get COVID, and will but that rate is still higher than during any previous recession need to quarantine themselves and care for others. Keeping since the Great Depression. While many workers have been parents, caregivers, and those who get sick integrated with the able to return to work, millions of workers and businesses are labor force will be key to minimizing the scarring effects of still trying to assess whether or how they can recover.

7 The Hamilton Project • Brookings Author

Betsey Stevenson has held previous positions at Princeton University and at the University of Pennsylvania’s Wharton School. Professor of Public Policy and Economics, University of Michigan Dr. Stevenson is a labor economist who has published widely in leading economics journals about the labor market and Betsey Stevenson is a professor of public policy and economics the impact of public policies on outcomes both in the labor at the University of Michigan. She is also a faculty research market and for families as they adjust to changing labor associate at the National Bureau of Economic Research, a market opportunities. Her research explores women’s visiting associate professor of economics at the University of labor market experiences, the economic forces shaping the Sydney, a research fellow of the Centre for Economic Policy modern family, and how these labor market experiences and Research, a fellow of the Ifo Institute for Economic Research economic forces on the family influence each other. She is a in Munich, and serves on the executive committee of the columnist for Bloomberg View, and her analysis of economic American Economic Association. She served as a member of data and the economy are frequently covered in both print the Council of Economic Advisers from 2013 to 2015 where and television media. she advised President Obama on social policy, labor market, and trade issues. She served as the chief economist of the Dr. Stevenson earned a B.A. in economics and mathematics U.S. Department of Labor from 2010 to 2011, advising the from Wellesley College and an M.A. and Ph.D. in economics Secretary of Labor on labor policy and participating as the from Harvard University. secretary’s deputy to the White House economic team. She

8 The Hamilton Project • Brookings Endnotes

1. There were 31.2 million initial claims for unemployment insurance filed 6. Between December 2007 and June 2009, men lost 2.6 jobs for every job lost between March 1 and May 2, not seasonally adjusted. In comparison, by women (Wood 2014). 1.8 million claims were filed over a similar period in 2019. Bartik et al. 7. See Petrongolo and Ronchi (2020) for a discussion of the ways in which both (2020) find that nearly all of the decline in hours worked occurred between the structure of the labor market and the choices women make contribute March 14 and March 28. Cajner et al. (2020) find that aggregate employment to the gender wage gap. See Goldin and Katz (2011) for discussion about the fell by 21 percent through late April before beginning to rebound. choices women make regarding workplace flexibility. 2. The Bureau of Labor Statistics (BLS) reported that the unemployment 8. This approach follows those respondents in waves 1 and 5 of the monthly rate was 19.2 percent if a correction is made for workers misclassified as Current Population Survey (CPS) in February; each respondent was employed but who were absent from work for other reasons (BLS 2020a). interviewed for four consecutive months from February to May. Additionally, Jason Furman and Wilson Powell III calculated a realistic 9. For a complete discussion of the misclassification of unemployed workers unemployment rate that peaked at 20.5 percent. They adjust both for the into the category, “employed but absent from work due to other reasons,” misclassification and for some of the reduction in labor force participation see Bauer et al. (2020). (Furman and Powell 2020). 10. Those other measures of nonemployment help explain the discrepancy 3. Goolsbee and Syverson (2020) find that legal restrictions account for between the larger number of people receiving unemployment insurance only 7 percentage points of the 60-percentage-point decline in consumer benefits and those measured as unemployed in the labor force statistics. traffic. Kahn, Lange, and Wiczer (2020) find that job postings deteriorated For example, in the week including May 12—the survey reference week substantially across the board, rather than more in states with shutdown for the CPS—a total of 30 million people were receiving unemployment orders. Similarly, Rojas et al. (2020) find that unemployment insurance insurance, but the CPS counted 21 million unemployed using the official claims rose across the board. unemployment rate. Accounting for some of those not in the labor force 4. Barrero, Bloom, and Davis (2020) examine the potential for sectoral shifts and employed but absent from work, as in Furman and Powell’s (2020) more and estimate that the pandemic will result in wide-scale labor market realistic unemployment rate, yields 27 million unemployed. Alternatively, reallocation. applying rates of nonemployment in May among those employed in 5. Women surpassed men on payrolls by 96,000 jobs in December 2019. February to the entire labor force would yield 33 million. Women first outnumbered men in nonfarm payrolls jobs in the second half 11. For more on how technological change is impacting the labor market, see of 2009. During this period, women held the majority of nonfarm payroll Autor and Reynolds (2020). jobs because men lost jobs faster than women did in the 2008 recession. However, employment among men recovered faster than employment among women; men once again held the majority of nonfarm payroll jobs by May 2010.

9 The Hamilton Project • Brookings References

Autor, David, and Elisabeth Reynolds. 2020. “The Nature of Work Furman, Jason, and Wilson Powell III. 2020. “U.S. Unemployment after the COVID Crisis: Too Few Low-Wage Jobs.” Essay Rate Falls Again But Little Progress After Accounting For 2020–14, The Hamilton Project, Brookings Institution, Recalls From Temporary Layoffs.” Peterson Institute for Washington, DC. International Economics, Realtime Economics Issues Watch Barrero, José María, Nick Bloom, and Steven J. Davis. 2020. (blog). “COVID-19 Is Also a Reallocation Shock.” Conference Goldin, Claudia, and Lawrence F. Katz. 2011. “The Cost of Draft,Brookings Papers on Economic Activity, Brookings Workplace Flexibility for High-Powered Professionals.” Institution, Washington, DC. Annals of the American Academy of Political and Social Bartik, Alexander W., Marianne Bertrand, Feng Lin, Jesse Science 638 (1): 45–67. Rothstein, and Matthew Unrath. 2020. “Measuring the Goolsbee, Austan, and Chad Syverson. 2020. “Fear, Lockdown, Labor Market at the Onset of the COVID-19 Crisis.” and Diversion: Comparing Drivers of Pandemic Economic Conference Draft,Brookings Papers on Economic Activity,, Decline 2020.” NBER Working Paper No. 27432, National Brookings Institution, Washington, DC. Bureau of Economic Research, Cambridge, MA. Bauer, Lauren, Wendy Edelberg, Jimmy O’Donnell, and Jay Kahn, Lisa B., Fabian Lange, and David G. Wiczer. 2020. “Labor Shambaugh. 2020. “Who Are the Potentially Misclassified Demand in the Time of COVID-19: Evidence from Vacancy in the Employment Report?” The Hamilton Project (blog). Postings and UI Claims.” NBER Working Paper No. 27061, Breitwieser, Audrey, Ryan Nunn, and Jay Shambaugh. 2018. “The National Bureau of Economic Research, Cambridge, MA. Recent Rebound in Prime-Age Labor Force Participation.” Petrongolo, Barbara, and Maddalena Ronchi. 2020. “Gender Gaps Brookings Up Front (blog). and the Structure of Local Labor Markets.” IZA Discussion Bureau of Labor Statistics [BLS]. 2000–20. “Labor Force Paper No. 13143, Institute of Labor Economics (IZA), Participation Rate – By Race and Ethnicity.” Bureau of Bonn, Germany. Labor Statistics, U.S. Department of Labor, Washington, Rojas, Felipe Lozano, Xuan Jiang, Laura Montenovo, Kosali I. DC. Retrieved from FRED, Federal Reserve Bank of St. Simon, Bruce A. Weinberg, and Coady Wing 2020. “Is the Louis. Cure Worse than the Problem Itself? Immediate Labor ———. 2020a. Frequently Asked Questions: The Impact of the Market Effects of COVID-19 Case Rates and School Coronavirus (COVID-19) Pandemic on The Employment Closures in the U.S.” NBER Working Paper No. 27127, Situation for April 2020. Washington, DC: U.S. Department National Bureau of Economic Research, Cambridge, MA. of Labor. Wood, Catherine A. 2014. “The Rise in Women’s Share of Nonfarm ———. 2020b. “Current Population Survey.” Bureau of Labor Employment During the 2007–2009 Recession: A Historical Statistics, U.S. Department of Labor, Washington, DC. Perspective.” Article, Monthly Labor Review, U.S. Bureau Retrieved from IPUMS-CPS, University of Minnesota. of Labor Statistics, U.S. Department of Labor, Washington, Cajner, Tomaz, Leland D. Crane Ryan A. Decker, John Grigsby, D.C. Adrian Hamins-Puertolas, Erik Hurst, Christopher Kurz, and Ahu Yildirmaz. 2020. “The U.S. Labor Market During the Beginning of the Pandemic Recession.” Conference Draft,Brookings Papers on Economic Activity, Brookings Institution, Washington, DC.

10 The Hamilton Project • Brookings ADVISORY COUNCIL

STEPHANIE AARONSON TIMOTHY F. GEITHNER ROBERT D. REISCHAUER Vice President and Director, Economic Studies President, Warburg Pincus Distinguished Institute Fellow & Senior Fellow, Economic Studies, The Brookings Senior Counselor, The Hamilton Project President Emeritus Institution Urban Institute JOHN GRAY GEORGE A. AKERLOF President & Chief Operating Officer NANCY L. ROSE University Professor Blackstone Charles P. Kindleberger Professor of Applied Georgetown University Economics, MIT Department of Economics ROBERT GREENSTEIN ROGER C. ALTMAN Founder & President DAVID M. RUBENSTEIN Founder & Senior Chairman Center on Budget and Policy Priorities Co-Founder & Co-Executive Chairman Evercore The Carlyle Group MICHAEL GREENSTONE KAREN L. ANDERSON Milton Friedman Professor in Economics & the ROBERT E. RUBIN Senior Director of Policy & Communications College Former U.S. Treasury Secretary; Becker Friedman Institute for Director of the Becker Friedman Institute for Co-Chair Emeritus Research in Economics Research in Economics Council on Foreign Relations The University of Chicago Director of the Energy Policy Institute University of Chicago LESLIE B. SAMUELS ALAN S. BLINDER Senior Counsel Gordon S. Rentschler Memorial Professor of GLENN H. HUTCHINS Cleary Gottlieb Steen & Hamilton LLP Economics & Public Affairs, Co-founder, North Island; Princeton University; Co-founder, Silver Lake SHERYL SANDBERG Visiting Senior Fellow, Chief Operating Officer, Facebook The Brookings Institution. JAMES A. JOHNSON Chairman; Johnson Capital Partners DIANE WHITMORE SCHANZENBACH STEVEN A. DENNING Margaret Walker Alexander Professor Chairman, General Atlantic LAWRENCE F. KATZ Director Elisabeth Allison Professor of Economics The Institute for Policy Research JOHN M. DEUTCH Harvard University Northwestern University; Institute Professor Nonresident Senior Fellow Massachusetts Institute of Technology MELISSA S. KEARNEY The Brookings Institution Neil Moskowitz Professor of Economics CHRISTOPHER EDLEY, JR. University of Maryland; STEPHEN SCHERR Co-Founder and President Emeritus Nonresident Senior Fellow Chief Executive Officer The Opportunity Institute The Brookings Institution Goldman Sachs Bank USA

BLAIR W. EFFRON LILI LYNTON RALPH L. SCHLOSSTEIN Partner Founding Partner President & Chief Executive Officer, Evercore Centerview Partners LLC Boulud Restaurant Group ERIC SCHMIDT DOUGLAS W. ELMENDORF HOWARD S. MARKS Technical Advisor, Alphabet Inc. Dean & Don K. Price Professor Co-Chairman of Public Policy Oaktree Capital Management, L.P. ERIC SCHWARTZ Harvard Kennedy School Chairman & CEO, 76 West Holdings ERIC MINDICH JUDY FEDER Founder THOMAS F. STEYER Professor & Former Dean Everblue Management Business Leader & Philanthropist McCourt School of Public Policy Georgetown University SUZANNE NORA JOHNSON MICHAEL R. STRAIN Former Vice Chairman Director of Economy Poliyc Studies and JASON FURMAN Goldman Sachs Group, Inc. Arthur F. Burns Scholar in Political Economy Professor of the Practice of Co-Chair American Enterprise Institute Economic Policy The Brookings Institution Harvard University LAWRENCE H. SUMMERS Senior Fellow PETER ORSZAG Charles W. Eliot University Professor Peterson Institute for International Economics; CEO, Financial Advisory Harvard University Senior Counselor Lazard Freres & Co LLC The Hamilton Project LAURA D’ANDREA TYSON RICHARD PERRY Distinguished Professor fo the Graduate School MARK T. GALLOGLY Managing Partner & Chief Executive Officer University of California, Berkeley Cofounder & Managing Principal Perry Capital Centerbridge Partners, L.P. PENNY PRITZKER WENDY EDELBERG TED GAYER Chairman & Founder, PSP Partners Director Executive Vice President 38th Secretary of Commerce Senior Fellow, Economic Studies JAY SHAMBAUGH The Brookings Institution MEEGHAN PRUNTY Director Managing Director, Blue Meridian Partners Edna McConnell Clark Foundation

11 The Hamilton Project • Brookings Summary

The economic damages of the COVID-19 pandemic are not being well captured by current labor market statistics that show both permanent damage to employment relationships and labor force attachment and the surge of workers who have experienced a temporary loss of work and income. The challenge is assessing the permanent damage that will persist well after the pandemic is behind us as a country. While the unemployment rate declined in May and June, permanent job loss accelerated over this period. Reversing this trend and getting these folks back to work is the difficult task that lies ahead of us. Just as the Great Recession disrupted the labor market attachment of millions of workers, a similar phenomenon will likely occur as a result of this recession. In this essay, Betsey Stevenson of the University of Michigan explores the many ways the COVID-19 recession has affected the labor market. Stevenson shows that the labor market effects have not been evenly borne across workers of different genders, races, and educational attainment. The scarring effects of the recession will likely lead to high long-term unemployment and weakened labor market attachment for years to come.

1775 Massachusetts Ave., NW Washington, DC 20036

(202) 797-6484

12 Printed on recycled paper. WWW.HAMILTONPROJECT.ORGThe Hamilton Project • Brookings